This dataset is accessed from https://www.kaggle.com/jessemostipak/college-tuition-diversity-and-... more This dataset is accessed from https://www.kaggle.com/jessemostipak/college-tuition-diversity-and-pay and was downloaded on August 4, 2021. <br><br> The following excerpt is from Kaggle regarding the sources of this dataset: The data this week comes from many different sources but originally came from the US Department of Education. Tuition and fees by college/university for 2018-2019, along with school type, degree length, state, in-state vs out-of-state from the Chronicle of Higher Education. Diversity by college/university for 2014, along with school type, degree length, state, in-state vs out-of-state from the Chronicle of Higher Education. Example diversity graphics from Priceonomics. Average net cost by income bracket from TuitionTracker.org. Example price trend and graduation rates from TuitionTracker.org Salary potential data comes from payscale.com. This dataset included the following files: 1. diversity_school.csv 2. historical_tuition.csv 3. salary_potential.cs...
The police shootings data is from the database maintained by Washington Post at https://github.co... more The police shootings data is from the database maintained by Washington Post at https://github.com/washingtonpost/data-police-shootings that covers the period 2015-2019.<br>The curated datasets are included here along with a research question and guiding questions.
The EJSCREEN environmental justice dataset is from EPA at https://www.epa.gov/ejscreen/download-e... more The EJSCREEN environmental justice dataset is from EPA at https://www.epa.gov/ejscreen/download-ejscreen-data.<br>Here we focus on the datasets corresponding to six states (Indiana, Louisiana, Iowa, Florida, New Hampshire and Vermont) for the year 2019.<br><br>The curated datasets are included here along with a research question and guiding questions.<br>See the codebook for full details.
Cervical cancer is the third most common cancer in women worldwide and the leading cause of cance... more Cervical cancer is the third most common cancer in women worldwide and the leading cause of cancer death in women of the developing countries. Cancer death rate can be greatly reduced by regular screening. One of the steps during a screening program is the detection of the abnormal cells that could evolve into cancer. In this paper, we propose an algorithm that automatically identifies the abnormal cervical regions from colposcopy image sequence. Firstly, based on the segmentation of three different image regions, a set of low-level features is extracted to model the temporal changes in the cervix before and after applying acetic acid. Second, a support vector machine (SVM) classifier is trained and used to make predictions on new input feature vectors. As the low-level features are very insensitive to accurate image registration, only a rough normalization step is needed to sample image patches. Our preliminary results show that our algorithm is accurate and effective. Furthermore,...
MILCOM 2015 - 2015 IEEE Military Communications Conference, 2015
The recent technological advances in mobile devices such as smartphones foster a wide variety of ... more The recent technological advances in mobile devices such as smartphones foster a wide variety of emerging applications which consider users as the providers as well as consumers of the highly valuable real world data from the devices. Interestingly enough, many of the existing researches related to this topic implicitly assume that the users will actively provide mobile sensing data to enable the applications without any compensation, which is not necessarily true for many reasons, e.g. extend battery lifetime, improve system performance, etc., and thus many users are rather dormant in practice. Therefore, there is an urgent need to develop a proper incentive mechanism for the applications to transform the users to be more active so that the applications can collect much-needed high-quality data. One common key challenge to realize the incentive mechanisms is how to preserve the privacy of the users as they will be requested to provide possibly-privacy-invasive mobile data. An anonymous identity and pseudonym based scheme is a straightforward and easy-to-adopt solution to address this issue. Unfortunately, this approach makes it extremely difficult or inefficient to detect duplicated sensing data from greedy users hoping to get more incentive with the state-of-art strategies. The duplicate data can generate lots of noise when the respective application analyzes the data and will cause more cost to operate the application, and therefore is very harmful. This paper proposes a novel privacy-aware mobile incentive scheme of its kind without trusted third party (TTP) in the sense that two different messages, each of which is with the same sensing data, but with different pseudonym, from the same mobile user can be used to recover the private key of the user.
Innovative Algorithms and Techniques in Automation, Industrial Electronics and Telecommunications
In this paper, we propose a general network model, fuzzy logic network (FLN), and study its stabi... more In this paper, we propose a general network model, fuzzy logic network (FLN), and study its stability and convergence properties. The convergence property was first deduced theoretically. Then a random approach was adopted to simulate the convergence speed and steady-state properties for a variety of fuzzy logical functions. The simulation results show that MV logical function causes the system to be on the edge of chaos when the number of nodes increases. Thus this logical function is more useful to infer real complex networks, such as gene regulatory networks.
In this chapter, a novel gene regulatory network inference algorithm based on the fuzzy logic net... more In this chapter, a novel gene regulatory network inference algorithm based on the fuzzy logic network theory is proposed and tested. The key motivation for this algorithm is that genes with regulatory relationships may be modeled via fuzzy logic, and the strength of regulations may be represented as the length of accumulated distance during a period of time intervals. One unique feature of this algorithm is that it makes very limited a priori assumptions concerning the modeling. Hence the algorithm is categorized as a data-driven algorithm. With the theoretical guidelines to quantify the upper limits of parameters, the algorithm is implemented to infer gene regulatory networks for Saccharomyces cerevisiae and Saccharomyces pombe. The computation results not only prove the validity of the data-driven algorithm, but also offer a possible explanation concerning the difference of network stabilities between the budding yeast and the fission yeast.
IEEE Transactions on Intelligent Transportation Systems, 2015
In this paper, we describe a novel algorithm that counts and classifies highway vehicles based on... more In this paper, we describe a novel algorithm that counts and classifies highway vehicles based on regression analysis. This algorithm requires no explicit segmentation or tracking of individual vehicles, which is usually an important part of many existing algorithms. Therefore, this algorithm is particularly useful when there are severe occlusions or vehicle resolution is low, in which extracted features are highly unreliable. There are mainly two contributions in our proposed algorithm. First, a warping method is developed to detect the foreground segments that contain unclassified vehicles. The common used modeling and tracking (e.g., Kalman filtering) of individual vehicles are not required. In order to reduce vehicle distortion caused by the foreshortening effect, a nonuniform mesh grid and a projective transformation are estimated and applied during the warping process. Second, we extract a set of low-level features for each foreground segment and develop a cascaded regression approach to count and classify vehicles directly, which has not been used in the area of intelligent transportation systems. Three different regressors are designed and evaluated. Experiments show that our regression-based algorithm is accurate and robust for poor quality videos, from which many existing algorithms could fail to extract reliable features.
Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems
Abstract-This paper discusses an approach to mo-bile robot path planning which utilizes a wavefro... more Abstract-This paper discusses an approach to mo-bile robot path planning which utilizes a wavefront-like propagation method for determining near-optimal paths. This approach is different from previous wave propagation methods in three aspects: it attempts to find near-optimal ...
2014 10th International Conference on Mobile Ad-hoc and Sensor Networks, 2014
The term "smart grid" refers to the next generation power supply system. A smart meter,... more The term "smart grid" refers to the next generation power supply system. A smart meter, an essential component of the grid system, is installed at each housing unit and acts as an agent for the unit. While the smart meter is a key enabler of great opportunities and conveniences in smart grid, it is susceptible to various cyber-security attacks, especially privacy invasion from electricity providers. Trusted third party (TTP) and homomorphic encryption are two favorite tools to deal with this issue in the literature. Unfortunately, the use of TTP does not completely eliminate the privacy risk. On the other hand, the use of homomorphic encryption makes it harder for the providers to support various services whose demand can be highly diversified. In this paper, we introduce a drastically new approach to deal with the consumer privacy issue in smart grid. Our key idea is let each consumer to determine the frequency of the measurement report. In this way, each consumer can responsibly make a trade-off between the level of privacy preservation with the quality of the services it will receive.
2014 22nd International Conference on Pattern Recognition, 2014
Iris segmentation is an important module of iris recognition that can substantially affect recogn... more Iris segmentation is an important module of iris recognition that can substantially affect recognition performance. Since iris and pupil boundaries usually are not exactly circular, spline-based methods have been used to model irregular iris and pupil boundaries recently. However, in most existing methods, many other factors or modules in the iris recognition pipeline are evaluated together and their mixed effects are assumed to be negligible. More importantly, the splines that model irregularity of the boundaries could not be enough to model the internal nonlinear deformations of an iris pattern (e.g., caused by iris dilation). As a result, it remains unclear whether spline-based methods can provide significant improvements. In this paper, we conduct a complete performance comparison between circular and splinebased methods. There are mainly two contributions. Firstly, for the purpose of comparison, we propose a spline estimator that is robust to outliers caused by eyelashes, eyelids, highlights, and shadows. Secondly, we analyze the relation between iris matching distances and segmentation results by using circular and splinebased methods. Based on our experiments, we found that, even with the proposed robust spline estimator, the improvement of recognition performance is still limited (around 6%). Therefore, in case that less robust spline estimators are used due to the real-time requirement in practical systems, the actual recognition improvement by using splines could be far below the expectation.
2014 IEEE International Conference on Multimedia and Expo (ICME), 2014
This paper presents a novel system framework of face beautification. Unlike prior works that deal... more This paper presents a novel system framework of face beautification. Unlike prior works that deal with single images, the proposed beautification framework is designed for an input video and it is able to improve both the appearance and the shape of a face. Our system adopts a state-of-the-art algorithm to synthesize and track 3D face models using blendshapes. The personalized 3D model can be edited to satisfy personal preference. This interactive process is needed only once per subject. Based on the tracking result and the modified face model, we present an algorithm to beautify the face video efficiently and consistently. Furthermore we develop a variant of content preserving warping to reduce warping distortions along the face boundary. Finally we adopt real time bilateral filtering to remove wrinkles, freckles, and unwanted blemishes. This framework is evaluated on a set of videos. The experiments demonstrate that our framework can generate consistent and pleasant results over video frames while the original expressions and features are persevered naturally.
In this chapter, a novel gene regulatory network gene regulatory network inference algorithm base... more In this chapter, a novel gene regulatory network gene regulatory network inference algorithm based on the fuzzy logic network fuzzy logic network is proposed and tested. The algorithm is intuitive and robust. The key motivation for this algorithm is that genes with regulatory relationships can be modeled via fuzzy logic, and the degrees of regulations can be represented as the
Visual Communications and Image Processing '90: Fifth in a Series, 1990
We present a method for generating shadows in a static environment represented by polygonal or pa... more We present a method for generating shadows in a static environment represented by polygonal or parametric surfaces, and illuminated by one or more movable point light sources. Concepts of ray tracing are modified and utilized in shadow determination. Ray-patch intersection are determined by using contour integration rather than purely numerical or subdivision techniques and thus the expense of long computation is somewhat lessened. Preprocessing steps are used to ascertain that the number of such calculations are limited. Shadow pairs are detected and pairs that cannot interact to produce shadows are also detected and discarded. Coherence properties are used to limit the number of rays processed. The approach combines elements of the shadow z-buffer in the generation process and to a lesser extent, that of projected polygons in scanline rendering as a preprocessing step. The algorithm is easily integrated with scanline a z-buffer algorithm and thus retains the benefits of a z- buffer. Distributed light sources are modelled as arrays of point light sources and the method allows the computation of umbra-penumbra effects.
This dataset is accessed from https://www.kaggle.com/jessemostipak/college-tuition-diversity-and-... more This dataset is accessed from https://www.kaggle.com/jessemostipak/college-tuition-diversity-and-pay and was downloaded on August 4, 2021. <br><br> The following excerpt is from Kaggle regarding the sources of this dataset: The data this week comes from many different sources but originally came from the US Department of Education. Tuition and fees by college/university for 2018-2019, along with school type, degree length, state, in-state vs out-of-state from the Chronicle of Higher Education. Diversity by college/university for 2014, along with school type, degree length, state, in-state vs out-of-state from the Chronicle of Higher Education. Example diversity graphics from Priceonomics. Average net cost by income bracket from TuitionTracker.org. Example price trend and graduation rates from TuitionTracker.org Salary potential data comes from payscale.com. This dataset included the following files: 1. diversity_school.csv 2. historical_tuition.csv 3. salary_potential.cs...
The police shootings data is from the database maintained by Washington Post at https://github.co... more The police shootings data is from the database maintained by Washington Post at https://github.com/washingtonpost/data-police-shootings that covers the period 2015-2019.<br>The curated datasets are included here along with a research question and guiding questions.
The EJSCREEN environmental justice dataset is from EPA at https://www.epa.gov/ejscreen/download-e... more The EJSCREEN environmental justice dataset is from EPA at https://www.epa.gov/ejscreen/download-ejscreen-data.<br>Here we focus on the datasets corresponding to six states (Indiana, Louisiana, Iowa, Florida, New Hampshire and Vermont) for the year 2019.<br><br>The curated datasets are included here along with a research question and guiding questions.<br>See the codebook for full details.
Cervical cancer is the third most common cancer in women worldwide and the leading cause of cance... more Cervical cancer is the third most common cancer in women worldwide and the leading cause of cancer death in women of the developing countries. Cancer death rate can be greatly reduced by regular screening. One of the steps during a screening program is the detection of the abnormal cells that could evolve into cancer. In this paper, we propose an algorithm that automatically identifies the abnormal cervical regions from colposcopy image sequence. Firstly, based on the segmentation of three different image regions, a set of low-level features is extracted to model the temporal changes in the cervix before and after applying acetic acid. Second, a support vector machine (SVM) classifier is trained and used to make predictions on new input feature vectors. As the low-level features are very insensitive to accurate image registration, only a rough normalization step is needed to sample image patches. Our preliminary results show that our algorithm is accurate and effective. Furthermore,...
MILCOM 2015 - 2015 IEEE Military Communications Conference, 2015
The recent technological advances in mobile devices such as smartphones foster a wide variety of ... more The recent technological advances in mobile devices such as smartphones foster a wide variety of emerging applications which consider users as the providers as well as consumers of the highly valuable real world data from the devices. Interestingly enough, many of the existing researches related to this topic implicitly assume that the users will actively provide mobile sensing data to enable the applications without any compensation, which is not necessarily true for many reasons, e.g. extend battery lifetime, improve system performance, etc., and thus many users are rather dormant in practice. Therefore, there is an urgent need to develop a proper incentive mechanism for the applications to transform the users to be more active so that the applications can collect much-needed high-quality data. One common key challenge to realize the incentive mechanisms is how to preserve the privacy of the users as they will be requested to provide possibly-privacy-invasive mobile data. An anonymous identity and pseudonym based scheme is a straightforward and easy-to-adopt solution to address this issue. Unfortunately, this approach makes it extremely difficult or inefficient to detect duplicated sensing data from greedy users hoping to get more incentive with the state-of-art strategies. The duplicate data can generate lots of noise when the respective application analyzes the data and will cause more cost to operate the application, and therefore is very harmful. This paper proposes a novel privacy-aware mobile incentive scheme of its kind without trusted third party (TTP) in the sense that two different messages, each of which is with the same sensing data, but with different pseudonym, from the same mobile user can be used to recover the private key of the user.
Innovative Algorithms and Techniques in Automation, Industrial Electronics and Telecommunications
In this paper, we propose a general network model, fuzzy logic network (FLN), and study its stabi... more In this paper, we propose a general network model, fuzzy logic network (FLN), and study its stability and convergence properties. The convergence property was first deduced theoretically. Then a random approach was adopted to simulate the convergence speed and steady-state properties for a variety of fuzzy logical functions. The simulation results show that MV logical function causes the system to be on the edge of chaos when the number of nodes increases. Thus this logical function is more useful to infer real complex networks, such as gene regulatory networks.
In this chapter, a novel gene regulatory network inference algorithm based on the fuzzy logic net... more In this chapter, a novel gene regulatory network inference algorithm based on the fuzzy logic network theory is proposed and tested. The key motivation for this algorithm is that genes with regulatory relationships may be modeled via fuzzy logic, and the strength of regulations may be represented as the length of accumulated distance during a period of time intervals. One unique feature of this algorithm is that it makes very limited a priori assumptions concerning the modeling. Hence the algorithm is categorized as a data-driven algorithm. With the theoretical guidelines to quantify the upper limits of parameters, the algorithm is implemented to infer gene regulatory networks for Saccharomyces cerevisiae and Saccharomyces pombe. The computation results not only prove the validity of the data-driven algorithm, but also offer a possible explanation concerning the difference of network stabilities between the budding yeast and the fission yeast.
IEEE Transactions on Intelligent Transportation Systems, 2015
In this paper, we describe a novel algorithm that counts and classifies highway vehicles based on... more In this paper, we describe a novel algorithm that counts and classifies highway vehicles based on regression analysis. This algorithm requires no explicit segmentation or tracking of individual vehicles, which is usually an important part of many existing algorithms. Therefore, this algorithm is particularly useful when there are severe occlusions or vehicle resolution is low, in which extracted features are highly unreliable. There are mainly two contributions in our proposed algorithm. First, a warping method is developed to detect the foreground segments that contain unclassified vehicles. The common used modeling and tracking (e.g., Kalman filtering) of individual vehicles are not required. In order to reduce vehicle distortion caused by the foreshortening effect, a nonuniform mesh grid and a projective transformation are estimated and applied during the warping process. Second, we extract a set of low-level features for each foreground segment and develop a cascaded regression approach to count and classify vehicles directly, which has not been used in the area of intelligent transportation systems. Three different regressors are designed and evaluated. Experiments show that our regression-based algorithm is accurate and robust for poor quality videos, from which many existing algorithms could fail to extract reliable features.
Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems
Abstract-This paper discusses an approach to mo-bile robot path planning which utilizes a wavefro... more Abstract-This paper discusses an approach to mo-bile robot path planning which utilizes a wavefront-like propagation method for determining near-optimal paths. This approach is different from previous wave propagation methods in three aspects: it attempts to find near-optimal ...
2014 10th International Conference on Mobile Ad-hoc and Sensor Networks, 2014
The term "smart grid" refers to the next generation power supply system. A smart meter,... more The term "smart grid" refers to the next generation power supply system. A smart meter, an essential component of the grid system, is installed at each housing unit and acts as an agent for the unit. While the smart meter is a key enabler of great opportunities and conveniences in smart grid, it is susceptible to various cyber-security attacks, especially privacy invasion from electricity providers. Trusted third party (TTP) and homomorphic encryption are two favorite tools to deal with this issue in the literature. Unfortunately, the use of TTP does not completely eliminate the privacy risk. On the other hand, the use of homomorphic encryption makes it harder for the providers to support various services whose demand can be highly diversified. In this paper, we introduce a drastically new approach to deal with the consumer privacy issue in smart grid. Our key idea is let each consumer to determine the frequency of the measurement report. In this way, each consumer can responsibly make a trade-off between the level of privacy preservation with the quality of the services it will receive.
2014 22nd International Conference on Pattern Recognition, 2014
Iris segmentation is an important module of iris recognition that can substantially affect recogn... more Iris segmentation is an important module of iris recognition that can substantially affect recognition performance. Since iris and pupil boundaries usually are not exactly circular, spline-based methods have been used to model irregular iris and pupil boundaries recently. However, in most existing methods, many other factors or modules in the iris recognition pipeline are evaluated together and their mixed effects are assumed to be negligible. More importantly, the splines that model irregularity of the boundaries could not be enough to model the internal nonlinear deformations of an iris pattern (e.g., caused by iris dilation). As a result, it remains unclear whether spline-based methods can provide significant improvements. In this paper, we conduct a complete performance comparison between circular and splinebased methods. There are mainly two contributions. Firstly, for the purpose of comparison, we propose a spline estimator that is robust to outliers caused by eyelashes, eyelids, highlights, and shadows. Secondly, we analyze the relation between iris matching distances and segmentation results by using circular and splinebased methods. Based on our experiments, we found that, even with the proposed robust spline estimator, the improvement of recognition performance is still limited (around 6%). Therefore, in case that less robust spline estimators are used due to the real-time requirement in practical systems, the actual recognition improvement by using splines could be far below the expectation.
2014 IEEE International Conference on Multimedia and Expo (ICME), 2014
This paper presents a novel system framework of face beautification. Unlike prior works that deal... more This paper presents a novel system framework of face beautification. Unlike prior works that deal with single images, the proposed beautification framework is designed for an input video and it is able to improve both the appearance and the shape of a face. Our system adopts a state-of-the-art algorithm to synthesize and track 3D face models using blendshapes. The personalized 3D model can be edited to satisfy personal preference. This interactive process is needed only once per subject. Based on the tracking result and the modified face model, we present an algorithm to beautify the face video efficiently and consistently. Furthermore we develop a variant of content preserving warping to reduce warping distortions along the face boundary. Finally we adopt real time bilateral filtering to remove wrinkles, freckles, and unwanted blemishes. This framework is evaluated on a set of videos. The experiments demonstrate that our framework can generate consistent and pleasant results over video frames while the original expressions and features are persevered naturally.
In this chapter, a novel gene regulatory network gene regulatory network inference algorithm base... more In this chapter, a novel gene regulatory network gene regulatory network inference algorithm based on the fuzzy logic network fuzzy logic network is proposed and tested. The algorithm is intuitive and robust. The key motivation for this algorithm is that genes with regulatory relationships can be modeled via fuzzy logic, and the degrees of regulations can be represented as the
Visual Communications and Image Processing '90: Fifth in a Series, 1990
We present a method for generating shadows in a static environment represented by polygonal or pa... more We present a method for generating shadows in a static environment represented by polygonal or parametric surfaces, and illuminated by one or more movable point light sources. Concepts of ray tracing are modified and utilized in shadow determination. Ray-patch intersection are determined by using contour integration rather than purely numerical or subdivision techniques and thus the expense of long computation is somewhat lessened. Preprocessing steps are used to ascertain that the number of such calculations are limited. Shadow pairs are detected and pairs that cannot interact to produce shadows are also detected and discarded. Coherence properties are used to limit the number of rays processed. The approach combines elements of the shadow z-buffer in the generation process and to a lesser extent, that of projected polygons in scanline rendering as a preprocessing step. The algorithm is easily integrated with scanline a z-buffer algorithm and thus retains the benefits of a z- buffer. Distributed light sources are modelled as arrays of point light sources and the method allows the computation of umbra-penumbra effects.
Uploads
Papers by Alade Tokuta